Oscar. Chang

Oscar. Chang
Yachay Tech · Department of Mathematics & Computation

PhD

About

47
Publications
35,336
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
295
Citations
Citations since 2017
23 Research Items
155 Citations
2017201820192020202120222023010203040
2017201820192020202120222023010203040
2017201820192020202120222023010203040
2017201820192020202120222023010203040
Introduction
Our current research deals with designing Neural Self-taught Agents, computer programs capable of exploring and learning new solutions by themselves with a combination of artificial neural nets, gradient descent and reinforcement learning. With these tools a multidisciplinary team has developed self-taught agents that drive spirited robots and are capable of learing to solve protein folding problems related to CORONAVIRUS control.
Additional affiliations
January 2017 - January 2021
Yachay Tech
Position
  • Professor
Description
  • Teaching and doing research in self-taught agents
May 2016 - present
IT-Empresarial
Position
  • Leader of System innovation
May 2016 - October 2016
IT - Empresarial
Position
  • Leader of Research & innovation

Publications

Publications (47)
Article
Full-text available
This work proposes a deep neural network (DNN) that accomplishes the reliable visual recognition of a chosen object captured with a webcam and moving in a 3D space. Auto encoding and substitutional reality are used to train a shallow net until it achieves zero tracking error in a discrete ambient. This trained individual is set to work in a real wo...
Conference Paper
Full-text available
Abstract— This work proposes a deep neural network (DNN) algorithm that accomplishes consistent sales forecasting for weekly data of pharmaceutical products. The resultant time series is used to train with backpropagation, step by step, a DNN, where shallow nets face selected scenarios, with different space-time data considerations. In each step, b...
Conference Paper
Full-text available
Experiments with rodents in mazes demonstrate that, in addition to visual cues, spatial localization and olfactory sense play a key role in orientation, foraging and eventually survival. Simulation at some level and understanding of this unique behavior is important for solving optimal routing problems. This article proposes a Reinforcement Learnin...
Article
Full-text available
Cocoa is a very important export product for many countries, including Ecuador. In terms of bean quality, international competition is fierce, especially in the control of so-called "diseased beans" as these can damage the flavor and quality of a whole set of beans and their combination makes grading a difficult task. Researchers, on the other hand...
Conference Paper
Full-text available
Protect the information has always been important concerns for society, and mainly now in digital era. Currently exists different platforms to manage critical and sensitive information, ranging from bank accounts to social media. All platforms have taken steps to guarantee that the data passing through them is protected from hackers. An essential s...
Chapter
According to the World Health Organization (WHO) cervix cancer is a real threat for women at earthly level. A practice to avoid those losses is an early diagnosis of the disease, generally done with the Papanicolaou or Pap test. This requires for a pathologist to check pap smear images in an arduous assignment, to determine the existence of suspici...
Article
Full-text available
This paper presents a computer simulation of a virtual robot that behaves as a peptide chain of the Hemagglutinin-Esterase protein (HEs) from human coronavirus. The robot can learn efficient protein folding policies by itself and then use them to solve HEs folding episodes. The proposed robotic unfolded structure inhabits a dynamic environment and...
Preprint
Full-text available
Intrusion detection into computer networks has become one of the most important issues in cybersecurity. Attackers keep on researching and coding to discover new vulnerabilities to penetrate information security system. In consequence computer systems must be daily upgraded using up-to-date techniques to keep hackers at bay. This paper focuses on t...
Preprint
Full-text available
This paper presents a computer simulation of a virtual robot that behaves as a peptide chain of the Hemagglutinin-Esterase protein (HEs) from human coronavirus. The robot can learn efficient protein folding policies by itself and then use them to solve HEs folding episodes. The proposed robotic unfolded structure inhabits a dynamic environment and...
Conference Paper
Full-text available
Intrusion detection into computer networks has become one of the most important issues in cybersecurity. Attackers keep on researching and coding to discover new vulnerabilities to penetrate information security system. In consequence computer systems must be daily upgraded using up-to-date techniques to keep hackers at bay. This paper focuses on t...
Conference Paper
Full-text available
Programmers with movement disorders do not currently have a language that aids them to write code. This work proposes the creation of E-Move, a friendly Domain-Specific Language (DSL) that tolerates involuntary typing errors. E-Move targets programmers who suffer from involuntary movements in their upper extremities related to movement disorders ca...
Conference Paper
Full-text available
This paper presents a biological inspired robot capable of learn by itself high level Tic-Tac-Toe playing policies and then use this knowledge to advantageously compete with humans. The robot comprises a robotic arm, an artificial vision system and a self-motivated neural agent which has the capability to explore in a simulated ambient, new forms o...
Conference Paper
Full-text available
In the Tic-Tac-Toe game community brilliant children and some adults discover, through playing experience, exceptional game situations where the current player declares him/herself a winner, no matter what the opponent does in the next moves. This paper proposes a Tic-Tac-Toe learning environment based on a self-motivated neural agent that learns b...
Conference Paper
Full-text available
All companies need an effective method to predict future sales, and several classic statistical methods exist and are heavily used in the industry. This work proposes a novel sales prediction method based on Convolutional Neural Networks. This type of neural network is generally used for image processing tasks. But in this work, we explore new appl...
Conference Paper
Full-text available
Sales forecast is a key issue in pharmacy wholesales and has a direct effect in inventory cost, it also represents a difficult problem, affected by factors like promotions, price changes and seasonal preferences. Traditional sales forecast techniques use historical sales data and recur-sive formulas, which limit their accuracy. More recently learni...
Conference Paper
Full-text available
Cryptography aims to make information unintelligible (en-cryption), as well as, to retrieve the original data (decryption). Good cryptography means that the information is encrypted in such a way that a brute force attack on the key or cryptography algorithm are all impossible. Up to date, several ciphers have been proposed. However, their vulnerab...
Article
Full-text available
The Consumer Price Index (CPI) is one of the most important economic indicators for countries’ characterization and is typically considered an official measure of inflation. The CPI considers the monthly price variation of a determined set of goods and services in a specific region, and it is key in the economic and social planning of a given count...
Conference Paper
Full-text available
Every year, billions of dollars are lost due to credit card fraud, causing huge losses for users and the financial industry. This kind of illicit activity is perhaps the most common and the one that causes most concerns in the finance world. In recent years great attention has been paid to the search for techniques to avoid this significant loss of...
Conference Paper
Full-text available
An Autoencoder is an artificial neural network used for unsupervised learning and for dimensionality reduction. In this work, an Autoencoder has been used to encrypt and decrypt digital information. So, it is implemented to code and decode characters represented in an 8-bit format, which corresponds to the size of ASCII representation. The Back-pro...
Conference Paper
Full-text available
This paper deals with Brain-Inspired robot controllers, based on a special kind of artificial neural structures that burn "dark" energy to promote the self-motivated initiation of behaviors. We exploit this ambient to train a virtual multi-joint robot, with many moving parts, muscles and sensors distributed through the robot body, interacting with...
Chapter
Full-text available
This paper deals with Brain-Inspired robot controllers, based on a special kind of artificial neural structures that burn “dark” energy to promote the self-motivated initiation of behaviors. We exploit this ambient to train a virtual multi-joint robot, with many moving parts, muscles and sensors distributed through the robot body, interacting with...
Article
Full-text available
This work proposes a deep neural network (DNN) algorithm that accomplishes consistent sales forecasting for weekly data of pharmaceutical products. The resultant time series is used to train with backpropagation, step by step, a DNN, where shallow nets face selected scenarios, with different space-time data considerations. In each step, by using a...
Conference Paper
This paper presents a method used for detection of optic nerve in fundus digital images; for this purpose, initially there is a preprocessing and segmentation of digital images of fundus taken from databases Messidor and Stare in order to stand out the veins and blood vessels of ocular region. The processed image is used in an artificial neural net...
Article
The rapid advance of technology has caused for engineers who graduate a few years ago to have a definite shortcomings in the use, design and maintenance of modern industrial control systems. Such is the case of the wireless technology WirelessHART, which by their high performance and flexibility of use, is increasingly common in todays industry. To...
Article
This research presents novel artificial vision techniques applied to the detection of features for strawberries used in the food industry. For this purpose, a computer vision system based in arti#64257;cial neural networks is used, organized as a deep architecture and trained with noise compensated learning. This combination originates a strong net...
Conference Paper
This paper presents the implementation of a tool of low cost for the diagnosis of control valves, aimed at the identification and analysis of the performance of such elements and used them for the education in Engineering. It was implanted a HART network that allows the communication with the positioner of the valve, and it was also built a display...
Conference Paper
Full-text available
Engineering students majoring in instrumentation and control require training modules to realize practices in industrial processes and automatic control. Usually these modules are expensive and most are designed to control a single variable. In the present work we design and built a low-cost training module, which makes possible to realize labs pra...
Conference Paper
Full-text available
Abstract—This work proposes a deep ANN architecture which accomplishes the reliable visual classification of abnormal Pap smear cell. The system is driven by independent agents where the first agent consists of a three layer ANN pretrained to closely track a reticle pattern. This net participates in a local close loop that oscillates and produces u...
Conference Paper
We present an artificial vision based system that quickly and semi-automatically calibrates water meters. The system operates with an artificial neural net architecture which is able to visually read analog needles and dynamically track the float position of a lab rotameter. By using a closed water flow circuit powered by a low power centrifugal pu...
Conference Paper
Full-text available
Resumen— Presentamos un sistema de visión artificial el cual utiliza principios operativos detectados en el cerebro de los insectos; particularmente la comunicación con código depurado entre el Lóbulo Antenal (AL) y el Cuerpo de Mushroom (MB). El sistema mira al mundo a través de imágenes tomadas de cámaras web o de bases de datos. Rutinas OpenCV c...
Conference Paper
Full-text available
Resumen— Presentamos un robot visual cuyo controlador neural asociado desarrolla una percepción realista de “Affordances” u “Oportunidades de uso”. El controlador aplica principios detectados en el cerebro de los insectos; particularmente la comunicación usando código estabilizado-esparcido entre el Lóbulo Antenal (AL) y el Cuerpo de Mushroom (MB)....
Conference Paper
Full-text available
We present a visual robot whose associated neural controller develops a realistic perception of affordances. The controller uses known insect brain principles; particularly the time stabilized sparse code communication between the Antennal Lobe and the Mushroom Body. The robot perceives the world through a webcam and canny border openCV routines. S...
Conference Paper
Full-text available
An artificial vision system based upon known insect brain structures is presented. It reliably recognizes real world objects visualized through a web cam or read from databases, and utilizes neural agents that communicate through time stabilized sparse code. A three layer ANN is trained to track one reticle pattern. Once trained the net becomes a p...
Conference Paper
Full-text available
We have studied and developed the behavior of two specific neural processes, used for vehicle driving and path planning, in order to control mobile robots. Each processor is an independent agent defined by a neural network trained for a defined task. Through simulated evolution fully trained agents are encouraged to socialize by opening low bandwid...
Conference Paper
Full-text available
A neural behavior initiating agent (BIA) is proposed to integrate relevant compressed image information coming from others cooperating and specialized neural agents. Using this arrangement the problem of tracking and recognizing a moving icon has been solved by partitioning it into three simpler and separated tasks. Neural modules associated to tho...
Article
Full-text available
Some problems in the field of non-stationary and highly non-lineal sign processing, such as voice recognition, have not been solved by traditional neural network. This proposal for implementation and simulation of a dynamic sinapse is offered to have a background for the biological neuron, specifically depression and facilitation phenomena ocurred...
Conference Paper
The synthesis of radiation patterns of linear arrays using the Schelkunoff method and genetic algorithms (GAs) is presented. GAs permit the location of the roots on the complex w plane until the desired pattern is obtained. This technique is a powerful tool in the synthesis of complex radiation patterns
Conference Paper
The synthesis of the radiation pattern of linear antenna arrays is an interesting problem in radiating systems. When the synthesis of antennas is subject to many restrictions the problem is very complex and becomes difficult to solve by classical optimization techniques because they are vulnerable to local minima problems. One relatively novel tech...
Conference Paper
Genetic algorithms are used to synthesize the radiation pattern of a linear array subject to many restrictions. In particular, a dual beam radiation pattern with low sidelobes is presented. The results show that genetic algorithms are robust and can solve complex antenna problems
Article
A novel approach to low-cost microprocessor-based speech analysis and processing has been developed. Real-time Spanish vowel recognition accuracy of 98-76 per cent with an average time response of 200 ms in a noisy room is possible with very simple hardware and software. The voice-recognition method can be implemented in any commercially available...
Article
An eight-transistor static logic gate has been designed, which generates the functions OR, NOR, and, nand, xor, and XNOR of two logic variables, under the dynamic control of two programming variables. In addition to its versatility as a building block, the dynamic programmability of the gate makes it a powerful tool for the efficient design of comp...
Article
An eight-transistor static logic gate has been designed, which generates the functions OR, NOR, AND, NAND, XOR, and XNOR of two logic variables, under the dynamic control of two programming variables. In addition to its versatility as a building block, the dynamic programmability of the gate makes it a powerful tool for the efficient design of comp...
Article
Thesis (M.S.)--Pennsylvania State University. Library holds archival microfilm negative and service copy,

Network

Cited By